Skip to content

Hailo Model Zoo v2.14

Latest
Latest
Compare
Choose a tag to compare
@HailoModelZoo HailoModelZoo released this 14 Jul 11:56
· 4 commits to master since this release
v2.14
96ae8c5
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.

Changelog

v2.14

  • Update to use Dataflow Compiler v3.30.0 (developer-zone <https://hailo.ai/developer-zone/>_)

  • Update to use HailoRT 4.20.0 (developer-zone <https://hailo.ai/developer-zone/>_)

  • New cascade API (experimental)

    • Currently supports PETRv2, bird-eye-view network for 3D object detection, see petrv2_repvggB0.yaml for configurations.

    • The user needs existing hars/hefs: both petrv2_repvggB0_backbone_pp_800x320 & petrv2_repvggB0_transformer_pp_800x320

    • full_precision evaluation: hailomz cascade eval petrv2

    • hardware evaluation: hailomz cascade eval petrv2 --override target=hardware

  • New task:

    • Human Action Recognition

      • Added support for (partial) Kinetics-400 dataset

      • Added r3d_18 to support this task

  • New Models:

    • YOLOv11 <https://arxiv.org/pdf/2410.17725>_ - nano, small, medium, large, x-large - Latest YOLO detectors
    • CLIP <https://arxiv.org/pdf/2103.00020>_ ViT-Large-14-Laion2B - Contrastive Language-Image Pre-training model [H15H and H10H only]
    • SWIN <https://arxiv.org/pdf/2103.14030>_ - tiny, small - Shifted-Windows Transformer based classification model
    • DaViT <https://arxiv.org/pdf/2204.03645>_ - tiny - Dual Attention Vision Transformer classification model [H15H and H10H only]
    • LeViT <https://arxiv.org/pdf/2104.01136>_ - levit128, levit192, levit384 - Transformer based classification model
    • EfficientFormer <https://arxiv.org/pdf/2212.08059>_ - l1 - Transformer based classification model
    • Real-ESRGAN <https://arxiv.org/pdf/2107.10833>_ - x2 - Super Resolution model
    • R3D_18 <https://pytorch.org/vision/stable/models.html#video-classification>_ - r3d_18 - Video Classification network for Human Action Recognition [H8 only]

  • Bug fixes

v2.13

  • Update to use Dataflow Compiler v3.29.0 (developer-zone <https://hailo.ai/developer-zone/>_)

  • Update to use HailoRT 4.19.0 (developer-zone <https://hailo.ai/developer-zone/>_)

  • Using jit_compile which reduces dramatically the emulation inference time of the Hailo Model Zoo models.

  • New tasks:

    • BEV: Multi-View 3D Object Detection

      • Added support for NuScenes dataset

      • Added PETRv2 with the following configuration:

        1. Backbone: RepVGG-B0 (800x320 input resolution)

        2. Transformer: 3 decoder layers, detection queries=304, replaced LN with UN

  • New Models:

    • CAS-ViT <https://arxiv.org/pdf/2408.03703>_ - S, M, T - Convolutional-Attention based classification model
    • YOLOv10 <https://arxiv.org/pdf/2405.14458>_ - base, x-large - Latest YOLO detectors
    • CLIP <https://arxiv.org/pdf/2103.00020>_ Text Encoders - ResNet50x4, ViT-Large

  • New retraining Docker containers for:

    • PETR - Multi-View 3D Object Detection

  • Introduced new flags for hailomz CLI:

    • --ap-per-class for measuring average-precision per-class. Relevant for object detection and instance segmentation tasks.

  • Bug fixes

Assets 2
Loading